MLFLOW · CVE-2026-0545 · AI agents & MCP
NO FIX YET
MLflow's FastAPI job endpoints under /ajax-api/3.0/jobs/ do not check authentication or authorization when basic-auth is enabled. With job execution turned on and a job function allowlisted, an unauthenticated network client can submit, read, search and cancel jobs, which can lead to remote code execution if the allowed jobs run privileged actions.
first public Apr 3, 2026 · 1 public exploit · 2 sources
Check it with your coding agent
It reads this page and tells you if your code or app is affected.
Teams running a self-hosted MLflow tracking server with the basic-auth app enabled are exposed only if job execution is turned on and at least one job function is allowlisted. The job endpoints bypass basic-auth entirely, so a server with job execution off or no allowlisted jobs is not affected by the code-execution path. Managed or hosted MLflow setups that do not expose this server to the network can skip it.
Not affected if job-execution-disabled, managed-hosting, no-allowlisted-jobs
Who cares mlflow, python, self-hosted
pip show mlflow; affected versions are <= 3.10.1.MLFLOW_SERVER_ENABLE_JOB_EXECUTION=true in the server environment.basic-auth is the only protection.In mlflow/mlflow, the FastAPI job endpoints under /ajax-api/3.0/jobs/* are not protected by authentication or authorization when the basic-auth app is enabled. This vulnerability affects the latest version of the repository. If job execution is enabled (MLFLOW_SERVER_ENABLE_JOB_EXECUTION=true) and any job function is allowlisted, any network client can submit, read, search, and cancel jobs without credentials, bypassing basic-auth entirely. This can lead to unauthenticated remote code execution if allowed jobs perform privileged actions such as shell execution or filesystem changes. Even if jobs are deemed safe, this still constitutes an authentication bypass, potentially resulting in job spam, denial of service (DoS), or data exposure in job results.
Are you affected?
Ask from here. InfraStack checks the affected versions against your stack and answers in your inbox.
InfraStack checks the affected versions against your stack and answers in Slack, with the evidence and the fix.
InfraStack checks it against your stack and answers on your report, with the evidence and the fix.
No fix yet, per the sources.
GitHub Advisory Database, CC-BY 4.0 · NVD data is provided by NIST and is public domain. Posts quoted belong to their authors.